Utility of SOFA and Δ-SOFA scores for predicting outcome in critically ill patients from the emergency department

Abstract

Objective: The condition of critically ill patients in the emergency department (ED) varies from moment to moment. The aims of this study are to quantify sequential organ failure assessment (SOFA) and changes in SOFA scores over time and determine its prognostic impact.

Patients and methods: This is a prospective observational cohort study. We included 269 patients consecutively admitted to the ICU from the ED over 18 months. The SOFA scores at ED admission (ED-SOFA) and ICU admission (ICU-SOFA) were obtained. Relative changes in SOFA scores were calculated as follows: Δ-SOFA=ICU-SOFA-ED-SOFA. Patients were divided into two groups depending on the Δ-SOFA score: (a) Δ-SOFA=0-1; and (b) Δ-SOFA more than or equal to 2.

Results: The median ED-SOFA score was two points (interquartile range: 1-4.5) and the Δ-SOFA score was 2 points (interquartile range: 0-3). The Δ-SOFA score was more powerful (area under the curve: 0.81) than the ED-SOFA score (area under the curve: 0.75) in predicting hospital mortality. Sixteen (6%) patients had a Δ-SOFA score less than 0, 116 (43%) patients had a Δ-SOFA=0-1, and 137 (51%) patients had a Δ-SOFA of at least 2 points. The probability of being alive at hospital discharge was 51 and 86.5% in Δ-SOFA of at least 2 and Δ-SOFA=0-1 groups, respectively (P<0.001). Risk factors for an increase of two or more SOFA points were age, cirrhosis, a diagnosis of sepsis, and a prolonged ED stay.

Conclusion: SOFA and changes in the SOFA score over time are potentially useful tools for risk stratification when applied to critically ill patients admitted to ICUs from the ED.